A method of avoiding arbitrarily shaped obstacle for Car-like Robot

Abstract

In this paper, we propose a method of path-planning and path-following for Car-like Robot. The method uses a grid map to define any shape obstacle. The shortest path can be generated by searching the grid map. To generate continuous curvature path, we introduce clothoid curves. Assuming that the driving speed is constant, deciding the ideal angular velocity from path information is possible. We design a controller that combines an ideal input and feedback input. This controller minimize path following error. The effectiveness of this method is shown by simulation.